Single corneal endothelial cell function evaluation method based on cellular morphology

Abstract

The invention discloses a single corneal endothelial cell function evaluation method based on cellular morphology, and relates to the technical field of biomedical engineering and image processing. According to the method, normal corneal endothelial cell images are collected through a corneal endothelial microscope, cell contours are segmented after denoising and enhancement processing, finite element analysis software is imported to construct normal and decompensated single-layer full corneal endothelial finite element models, different pressures are applied to the models for simulation analysis, the relative change multiplying power of all parameter values is calculated, and the corneal endothelial cell contours are obtained. Obtaining the relationship between the Y-shaped gap area and the corneal endothelial cell function, constructing a model based on a simulation result, determining that the Y-shaped gap area can measure the cell function through inspection, and judging the corneal endothelial cell function state according to the relative change multiplying power threshold value of the Y-shaped gap area. The method can effectively evaluate the function of a single corneal endothelial cell, provides a more accurate basis for early diagnosis and treatment of corneal diseases, and assists the development of ophthalmology medical technology.

Description

Technical Field

[0001] The invention relates to the field of biomedical engineering and image processing technology, and in particular to a method for evaluating the function of a single corneal endothelial cell based on cell morphology. Background Art

[0002] Corneal endothelial cells play a key role in maintaining corneal transparency and regulating the hydration state of the corneal stroma. Their barrier and fluid pump functions are crucial. There are a large number of corneal blind patients in the world, about 12.7 million, of which corneal endothelial blindness accounts for 5%-10%, seriously affecting the patients' vision health. Accurately assessing the function of corneal endothelial cells is of great significance for the diagnosis and treatment of corneal diseases and the prevention of the occurrence of corneal endothelial blindness.

[0003] There are many limitations in the current clinical corneal endothelial detection methods. Although the slit lamp can indirectly determine whether the corneal endothelium is decompensated by observing whether the cornea is edematous, it lacks quantitative indicators, the diagnostic results are highly subjective, and it is difficult to accurately evaluate the endothelial function status. The corneal endothelial microscope can obtain corneal endothelial cell density, hexagon ratio and coefficient of variation indicators, which cannot directly and effectively reflect the endothelial function. In terms of in vitro cell detection, tissue staining, transmembrane resistance and Na+ / K+ATPase enzyme activity are used to evaluate cell barrier and pump functions, but they cannot directly reflect the true functional status of cells in vivo. In the field of ophthalmology, finite element analysis technology has been applied in the mechanical simulation of the cornea, eyeball and its various layers of tissue, but a mature technical system has not yet been formed. Therefore, how to achieve non-invasive, direct and accurate evaluation of the function of corneal endothelial cells has become a key issue that needs to be solved in the current field of corneal endothelial cell function detection. To this end, a single corneal endothelial cell function evaluation method based on cell morphology is proposed. Summary of the invention

[0004] The object of the present invention is to provide a method for evaluating the function of a single corneal endothelial cell based on cell morphology, so as to solve the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A method for evaluating the function of a single corneal endothelial cell based on cell morphology comprises the following steps:

[0007] S1. Collect images of normal corneal endothelial cells through corneal endothelial microscopy, perform denoising and enhancement on the images, segment the corneal endothelial cell contours using the global threshold method on the processed images, import the segmented corneal endothelial cell contours into the finite element analysis software ANSYS, construct a normal single-layer full corneal endothelial finite element model, adjust the beam unit parameters of the normal single-layer full corneal endothelial finite element model to generate a decompensated single-layer full corneal endothelial finite element model;

[0008] S2. Apply pressures of 10 mmHg, 15 mmHg, and 20 mmHg to the normal and decompensated single-layer full corneal endothelium finite element models respectively for simulation analysis. Taking the parameter values of corneal endothelial cells including the maximum displacement, edge stress, F / L ratio, and Y-shaped gap area at a pressure of 10 mmHg as the reference values, calculate the relative change multiples of each parameter value under the pressure conditions of 15 mmHg and 20 mmHg. Based on the relative change multiples of each parameter value of corneal endothelial cells in the normal and decompensated single-layer full corneal endothelium finite element models, obtain the simulation results related to the relationship between the Y-shaped gap area and the function of corneal endothelial cells;

[0009] S3. Based on the simulation results, select 5 images of corneal endothelial cells with clinically diagnosed corneal endothelial dysfunction and normal respectively, construct a single-layer full corneal endothelium finite element model, calculate the Y-shaped gap area, and through an independent samples t-test, it is obtained that the Y-shaped gap area of endothelial cells with corneal endothelial dysfunction is higher than that of endothelial cells with normal corneal endothelial function, and it is determined that the Y-shaped gap area can measure the function of corneal endothelial cells;

[0010] S4. Determine the threshold of the relative change multiple of the Y-shaped gap area based on the relative change multiple of the Y-shaped gap area of normal corneal endothelial cells, collect images of corneal endothelial cells, construct a single-layer full corneal endothelium finite element model, obtain the relative change multiple of the Y-shaped gap area, and thereby judge the functional state of corneal endothelial cells.

[0011] Preferably, a method for evaluating the function of a single corneal endothelial cell based on cell morphology is characterized in that the method for denoising and enhancing the image is as follows:

[0012] Perform denoising processing on the image through non-local means denoising. Collect images of normal corneal endothelial cells through a corneal endothelial microscope. For each pixel in the image, find a pixel block composed of multiple adjacent pixels similar to the neighborhood window of the pixel within the set search window through the Euclidean distance calculation formula. The similar pixel blocks have a certain similarity in gray-scale characteristics. Perform weighted average processing on the pixel values used to describe the gray-scale characteristics of the pixel block to obtain the denoised pixel, and generate the denoised image for all pixels of the denoised image;

[0013] The Euclidean distance calculation formula is:

[0014] where A and B are pixel blocks, a i is the pixel value at the corresponding position in pixel block A, b i is the pixel value at the corresponding position in pixel block B, d(A,B) is the Euclidean distance between pixel blocks A and B, and n is the number of pixels in the pixel block;

[0015] The denoised image is enhanced by contrast-limited adaptive histogram equalization (CLAHE). The denoised image is segmented into non-overlapping sub-blocks according to a set block size. For each sub-block, the pixel values are adjusted by calculating the cumulative distribution function (CDF) to enhance the contrast, making the details of the image more prominent. During the sub-block segmentation and contrast adjustment processes, discontinuities may occur at the edges of the sub-blocks. The bilinear interpolation method is used to interpolate the edges of adjacent sub-blocks to make the transition between adjacent sub-blocks natural. The individual sub-blocks are combined to obtain the enhanced image.

[0016] The method of adjusting pixel values by calculating the cumulative distribution function (CDF) to enhance the contrast is as follows:

[0017] The cumulative distribution function (CDF) is calculated using the CDF calculation formula. Based on the calculated cumulative distribution function, assuming the gray value of a pixel in the denoised image is x and the adjusted gray value is y, pixel value adjustment is achieved through a mapping formula. CLAHE limits the degree of contrast enhancement to prevent noise amplification or detail loss caused by over-enhancement. A contrast limit threshold T is set. If the histogram h(i) of a certain gray level i exceeds the threshold T, it is clipped to T, and the clipped histogram is denoted as h clip (i). The clipped histogram is renormalized so that According to the clipped histogram h clip (i), a new cumulative distribution function (CDF) is calculated using the CDF calculation formula, and the pixel values are re-adjusted accordingly.

[0018] The cumulative distribution function (CDF) calculation formula is:

[0019] where cdf(i) is the cumulative distribution function, h(j) is the gray histogram, and j and i are gray levels.

[0020] The mapping formula is: y = round((L - 1) × cdf(x));

[0021] where round() is the rounding function, x is the gray value of a pixel in the denoised image, y is the adjusted gray value, cdf(x) is the cumulative distribution function, and L is the total number of gray levels.

[0022] Preferably, the method of segmenting the corneal endothelial cell contour using the global threshold method for the processed image is:

[0023] The processed image is read by the cv2.imread() function in Python. The processed image is stored in the computer memory in the form of a matrix. Each element of the matrix corresponds to a pixel in the processed image, and the value of the element represents the grayscale information of the pixel. The processed image in the computer memory is generated by the cv2.calcHist() function of Python to generate an array containing 256 elements. Each element in the array corresponds to a grayscale level, and the array value represents the frequency of the grayscale level in the image. Based on this, a grayscale histogram is drawn by the matplotlib library of Python. The horizontal axis of the grayscale histogram represents the grayscale level, and the vertical axis represents the pixel frequency corresponding to the grayscale level. According to the peaks and troughs of the grayscale histogram, the global threshold is determined by the Otsu algorithm that automatically determines the threshold. Based on this, all pixels of the processed image are traversed, and pixels with grayscale values greater than the global threshold are determined as corneal endothelial cell pixels and the grayscale value is set to 0. Pixels with grayscale values less than the global threshold are determined as corneal endothelial cell contour pixels, and the grayscale value is set to 255. Based on this, the segmented corneal endothelial cell contour is obtained.

[0024] Preferably, the method for adjusting the beam unit parameters of the normal single-layer full corneal endothelial finite element model to generate the decompensated single-layer full corneal endothelial finite element model is:

[0025] The beam unit parameters are elastic modulus and cross-sectional area, which are used to simulate the mechanical behavior of corneal endothelial cells. In the corneal endothelial cell model, the elastic modulus represents the characteristics of intercellular junction proteins in maintaining their own shape and resisting deformation when subjected to force, and the cross-sectional area represents the performance of junction proteins in transmitting force and bearing load. Based on the existing research data on the mechanical properties of junction proteins between corneal endothelial cells, the elastic modulus E0 and cross-sectional area A0 of normal corneal endothelial junction proteins are determined. According to the characteristics of decreased stiffness of junction proteins during decompensation, the elastic modulus in the beam unit parameters of the normal single-layer full corneal endothelial finite element model is adjusted on the basis of the elastic modulus E0. The elastic modulus adjustment range is 0.2E0 to 0.5E0, which is used to simulate the changes in the elastic properties of junction proteins under decompensation. According to the characteristics of reduced number of junction proteins during decompensation, the cross-sectional area in the beam unit parameters of the normal single-layer full corneal endothelial finite element model is adjusted on the basis of the cross-sectional area A0. The cross-sectional area adjustment range is 0.5A0 to 0.8A0, which is used to reflect the mechanical changes of the intercellular connection structure in the decompensated state.

[0026] Preferably, the method for calculating the relative change ratio of each parameter value under the pressure conditions of 15 mmHg and 20 mmHg is as follows:

[0027] Using the finite element analysis software ANSYS, a pressure of 10 mmHg was applied to the normal and decompensated single-layer total corneal endothelium finite element models respectively, and the parameter values of the corneal endothelial cells such as the maximum displacement, edge stress, F / L ratio, and Y-shaped gap area were obtained. The parameter values were recorded as the reference values for calculating the relative change magnification. The maximum displacement of the normal single-layer total corneal endothelium finite element model under a pressure of 10 mmHg was D 10-ZC , the edge stress was S 10-ZC , the F / L ratio was R 10-ZC , and the Y-shaped gap area was A 10-ZC , the maximum displacement of the decompensated single-layer total corneal endothelium finite element model under a pressure of 10 mmHg was D 10-SDC , the edge stress was S 10-SDC , the F / L ratio was R 10-SDC , and the Y-shaped gap area was A 10-SDC ;

[0028] Using the finite element analysis software ANSYS, the parameter values of the corneal endothelial cells such as the maximum displacement, edge stress, F / L ratio, and Y-shaped gap area of the normal and decompensated single-layer total corneal endothelium finite element models under pressures of 15 mmHg and 20 mmHg were obtained. The maximum displacement of the normal single-layer total corneal endothelium finite element model under a pressure of 15 mmHg was D 15-ZC , the edge stress was S 15-ZC , the F / L ratio was R 15-ZC , and the Y-shaped gap area was A 15-ZC , the maximum displacement of the decompensated single-layer total corneal endothelium finite element model under a pressure of 15 mmHg was D 15-SDC , the edge stress was S 15-SDC , the F / L ratio was R 15-SDC , and the Y-shaped gap area was A 15-SDC ;

[0029] The maximum displacement of the normal single-layer total corneal endothelium finite element model under a pressure of 20 mmHg was D 20-ZC , the edge stress was S 20-ZC , the F / L ratio was R 20-ZC , and the Y-shaped gap area was A 20-ZC , the maximum displacement of the decompensated single-layer total corneal endothelium finite element model under a pressure of 20 mmHg was D 20-SDC , the edge stress was S 20-SDC , the F / L ratio was R 20-SDC , and the Y-shaped gap area was A 20-SDC ;

[0030] For the parameter values under the pressure conditions of 15 mmHg and 20 mmHg, through the relative change magnification calculation formula, that is the relative change magnification of each parameter value was obtained.

[0031] Preferably, the method for obtaining that the Y-shaped gap area is related to the corneal endothelial cell function based on the relative change multiples of the respective parameter values of the corneal endothelial cells in the normal and decompensated single-layer total corneal endothelium finite element models is as follows:

[0032] Divide the relative change multiple data of the respective parameter values of the normal and decompensated single-layer total corneal endothelium finite element models under 15 mmHg and 20 mmHg pressures into a normal group and a decompensated group. The normal group includes the relative change multiples of the respective parameter values of the normal single-layer total corneal endothelium finite element model under 15 mmHg and 20 mmHg pressures, and the decompensated group includes the relative change multiples of the respective parameter values of the decompensated single-layer total corneal endothelium finite element model under 15 mmHg and 20 mmHg pressures;

[0033] Analyze the trends of the relative change multiples of the respective parameter values with pressure changes in the normal group and the decompensated group, find the parameter value with the largest growth rate of the relative change multiple under 15 mmHg and 20 mmHg pressures. At the same time, if the relative change multiple of the parameter value with the largest growth rate in the decompensated group is higher than the relative change multiple in the corresponding normal group, it is concluded that this parameter value is related to the corneal endothelial cell function, and based on this, it is obtained that the Y-shaped gap area is related to the corneal endothelial cell function.

[0034] Preferably, the method for determining that the Y-shaped gap area of the endothelial cells with corneal endothelial dysfunction can be measured by the independent sample t-test, and it is determined that the Y-shaped gap area can measure the corneal endothelial cell function is as follows:

[0035] Input the Y-shaped gap areas corresponding to the images of the corneal endothelial cells with corneal endothelial dysfunction and normal function into the analysis software SPSS respectively, and perform statistical analysis through the independent sample t-test method, which is a statistical method for comparing whether the differences in the means of two independent samples are significant. According to the means, standard deviations, and sample sizes of the data of the Y-shaped gap areas corresponding to the images of the corneal endothelial cells with corneal endothelial dysfunction and normal function, obtain the t statistic, which is used to reflect the size of the mean difference of the data and the dispersion degree of the samples. The analysis software SPSS obtains that the P value corresponding to the Y-shaped gap area is less than the preset 0.05, then it is determined that the Y-shaped gap area can measure the corneal endothelial cell function.

[0036] Preferably, the method for determining the threshold of the relative change multiple of the Y-shaped gap area based on the relative change multiple of the Y-shaped gap area of normal corneal endothelial cells is as follows:

[0037] Collect 10 image samples of clinically normal corneal endothelial cells, construct 10 normal monolayer full corneal endothelial finite element model samples, and accordingly obtain the relative change magnification of the Y-shaped gap area of 10 normal corneal endothelial cells. Calculate the mean of the relative change magnification data for the relative change magnification of the Y-shaped gap area of 10 normal corneal endothelial cells. And the standard deviation s, and accordingly construct the relative change magnification threshold of the Y-shaped gap area;

[0038] The method for constructing the relative change magnification threshold of the Y-shaped gap area accordingly is as follows:

[0039] Calculate the mean of the relative change magnification data for the relative change magnification of the Y-shaped gap area of 10 normal corneal endothelial cells. And the standard deviation s, and substitute them into the designed threshold calculation method, that is, the mean Plus 2 times the standard deviation s to obtain the relative change magnification threshold of the Y-shaped gap area.

[0040] Preferably, the method for judging the functional state of corneal endothelial cells is as follows:

[0041] Compare the relative change magnification of the Y-shaped gap area of corneal endothelial cells with the relative change magnification threshold of the Y-shaped gap area;

[0042] If the relative change magnification of the Y-shaped gap area of corneal endothelial cells is less than the relative change magnification threshold of the Y-shaped gap area, it is judged that the function of corneal endothelial cells is in a normal state;

[0043] If the relative change magnification of the Y-shaped gap area of corneal endothelial cells is greater than or equal to the relative change magnification threshold of the Y-shaped gap area, it is judged that the function of corneal endothelial cells is abnormal.

[0044] Due to the adoption of the above technical solution, the technical progress achieved by the present invention compared with the prior art is:

[0045] 1. The present invention solves the limitations of the existing evaluation technology and realizes accurate and non-invasive quantitative evaluation. The present invention uses corneal endothelial cell images to construct a finite element model, and determines the relative change magnification threshold of the Y-shaped gap area by analyzing the relative change magnification of the maximum displacement, edge stress, F / L ratio and Y-shaped gap area under different pressures, realizing non-invasive and accurate quantitative evaluation of the function of corneal endothelial cells. In the embodiments of the present invention, according to the parameter changes of the model under different pressures, the functional state of cells is accurately judged, making up for the deficiencies of the prior art.

[0046] 2. The present invention expands the application of finite element technology in ophthalmology to facilitate the development of corneal disease diagnosis and treatment. The application of finite element analysis technology in ophthalmology mainly focuses on the mechanical simulation of the cornea, eyeball and their various tissue layers, and is less applied in the research of corneal endothelial cell function. The present invention innovatively uses finite element analysis to construct normal and decompensated models, deeply studies the function of corneal endothelial cells, and through clinical sample verification, clarifies the relationship between the Y-shaped gap area and cell function, providing a more accurate basis for the early diagnosis and treatment of corneal diseases.

[0047] Attached explanatory drawings

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0049] Figure 1 It is a schematic flowchart of the method of the present invention.

[0050] Specific embodiments

[0051] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0052] Example 1, as Figure 1 described, a method for evaluating the function of a single corneal endothelial cell based on cell morphology includes the following steps:

[0053] S1. Collect images of normal corneal endothelial cells through a corneal endothelial microscope, perform denoising and enhancement processing on the images, use the global threshold method to segment the corneal endothelial cell contours on the processed images, import the segmented corneal endothelial cell contours into the finite element analysis software ANSYS, construct a normal single-layer full corneal endothelial finite element model, and adjust the beam element parameters of the normal single-layer full corneal endothelial finite element model to generate a decompensated single-layer full corneal endothelial finite element model;

[0054] S2. Apply pressures of 10 mmHg, 15 mmHg, and 20 mmHg to the normal and decompensated monolayer full corneal endothelium finite element models respectively for simulation analysis. Taking the parameter values of corneal endothelial cells including the maximum displacement, edge stress, F / L ratio, and Y-shaped gap area at a pressure of 10 mmHg as the reference values, calculate the relative change multiples of each parameter value under the pressure conditions of 15 mmHg and 20 mmHg. Based on the relative change multiples of each parameter value of corneal endothelial cells in the normal and decompensated monolayer full corneal endothelium finite element models, obtain the simulation results related to the relationship between the Y-shaped gap area and the function of corneal endothelial cells;

[0055] S3. Based on the simulation results, select 5 images of corneal endothelial cells with clinically diagnosed corneal endothelial dysfunction and normal conditions respectively, construct a monolayer full corneal endothelium finite element model, calculate the Y-shaped gap area, and obtain through an independent samples t-test that the Y-shaped gap area of endothelial cells with corneal endothelial dysfunction is higher than that of endothelial cells with normal corneal endothelial function, and determine that the Y-shaped gap area can measure the function of corneal endothelial cells;

[0056] S4. Determine the threshold of the relative change multiple of the Y-shaped gap area based on the relative change multiple of the Y-shaped gap area of normal corneal endothelial cells, collect images of corneal endothelial cells, construct a monolayer full corneal endothelium finite element model, obtain the relative change multiple of the Y-shaped gap area, and thereby judge the functional state of corneal endothelial cells.

[0057] Furthermore, the working principle of the present invention is illustrated by the following embodiments:

[0058] Using a high-precision corneal endothelial microscope, images of normal corneal endothelial cells were obtained from healthy volunteers aged between 22 and 48 years old with no eye diseases. There was a certain degree of noise interference in the obtained images of normal corneal endothelial cells. The non-local means denoising method was used to process the images. The search window was set in the range of 30×30 pixels, and the neighborhood window was in the range of 10×10 pixels. Each pixel in the image was traversed, and similar pixel blocks were found through the Euclidean distance calculation formula. Then, the pixel values used to describe the pixel gray characteristics in these pixel blocks were weighted and averaged to obtain the denoised pixels. All the denoised pixels constituted the denoised image. For the denoised image, contrast-limited adaptive histogram equalization (CLAHE) was used for enhancement processing. The image was segmented into non-overlapping sub-blocks with a block size of 20×20 pixels. For each sub-block, the pixel values were adjusted by calculating the cumulative distribution function (CDF) to enhance the contrast. The bilinear interpolation method was used to interpolate the edges of adjacent sub-blocks. Finally, the sub-blocks were merged to obtain the enhanced image. The processed image was read using the cv2.imread() function in the cv2 library of Python. The image was stored in the computer memory in matrix form. An array containing 256 elements was generated using the cv2.calcHist() function, and then the gray histogram was plotted using the matplotlib library. According to the peaks and valleys of the gray histogram, the global threshold was determined by the Otsu algorithm. All pixels in the processed image were traversed. Pixels with gray values greater than the global threshold were determined as corneal endothelial cell pixels, and their gray values were set to 0. Pixels with gray values less than the global threshold were determined as corneal endothelial cell contour pixels, and the gray values were set to 255. In this way, the corneal endothelial cell contour was successfully segmented. The segmented corneal endothelial cell contour was imported into the finite element analysis software ANSYS to construct a normal single-layer full corneal endothelial finite element model. Referring to the existing research data on the mechanical properties of connexins between corneal endothelial cells, the elastic modulus E0 of connexins between normal corneal endothelial cells was determined to be 550 MPa, and the cross-sectional area A0 was 0.055 mm 2 , based on the characteristics of decreased stiffness and reduced quantity of connexins during decompensation, the beam element parameters of the normal single-layer full corneal endothelial finite element model were adjusted. The elastic modulus was adjusted in the range of 0.2E0 to 0.5E0, that is, 110 MPa to 275 MPa, and the cross-sectional area was adjusted in the range of 0.5A0 to 0.8A0, that is, 0.0275 mm 2 to 0.044 mm 2 to generate a decompensated single-layer full corneal endothelial finite element model.

[0059] Using the finite element analysis software ANSYS, simulations were performed on normal and decompensated single-layer total corneal endothelium finite element models by applying pressures of 10 mmHg, 15 mmHg, and 20 mmHg respectively. Taking the parameter values under 10 mmHg pressure as the reference values, the maximum displacement D of the normal single-layer total corneal endothelium finite element model under 10 mmHg pressure 10-ZC was 5.2 μm, and the edge stress S 10-ZC was 0.8 MPa, and the F / L ratio R 10-ZC was 0.6, and the Y-shaped gap area A 10-ZC was 20.5 μm 2 , and the maximum displacement D of the decompensated single-layer total corneal endothelium finite element model under 10 mmHg pressure 10-SDC was 7.5 μm, and the edge stress S 10-SDC was 1.1 MPa, and the F / L ratio R 10-SDC was 0.7, and the Y-shaped gap area A 10-SDC was 28.0. The maximum displacement D of the normal single-layer total corneal endothelium finite element model under 15 mmHg pressure 15-ZC was 6.8 μm, and the edge stress S 15-ZC was 1.2 MPa, and the F / L ratio R 15-ZC was 0.7, and the Y-shaped gap area A 15-ZC was 25.6 μm 2 , and the maximum displacement D of the decompensated single-layer total corneal endothelium finite element model under 15 mmHg pressure 15-SDC was 10.2 μm, and the edge stress S 15-SDC was 1.8 MPa, and the F / L ratio R 15-SDC was 0.9, and the Y-shaped gap area A 15-SDC was 40.5. The maximum displacement D of the normal single-layer total corneal endothelium finite element model under 20 mmHg pressure 15-ZC was 8.5 μm, and the edge stress S 15-ZC was 1.6 MPa, and the F / L ratio R 15-ZC was 0.8, and the Y-shaped gap area A 15-ZC was 32.0 μm 2 , and the maximum displacement D of the decompensated single-layer total corneal endothelium finite element model under 20 mmHg pressure 15-SDC was 13.5 μm, and the edge stress S 15-SDC was 2.5 MPa, and the F / L ratio R 15-SDC was 1.1, and the Y-shaped gap area A 15-SDC was 55.0 μm 2, according to the relative change magnification calculation formula, taking the Y-shaped gap area as an example, the relative change magnification of the Y-shaped gap area in the normal single-layer full corneal endothelial finite element model is about 1.24 at a pressure of 15 mmHg, and about 1.53 at a pressure of 20 mmHg. The relative change magnification of the Y-shaped gap area in the decompensated single-layer full corneal endothelial finite element model is about 1.38 at a pressure of 15 mmHg, and about 1.88 at a pressure of 20 mmHg. By comparing the relative change magnifications of each parameter at different pressures, it is found that in the decompensated model, as the pressure increases, the growth rate of the relative change magnification of the Y-shaped gap area is the largest, and it is significantly higher than the relative change magnification of the normal model under the same pressure change.

[0060] Corneal endothelial cell images of 5 patients clinically diagnosed with corneal endothelial decompensation and 5 individuals with normal corneal endothelial function were selected, and 5 single-layer full corneal endothelial finite element models were constructed respectively. The Y-shaped gap area was calculated, and the independent sample t-test method was used to statistically analyze the Y-shaped gap area data. The obtained P value corresponding to the Y-shaped gap area was 0.01, which was much smaller than the pre-set 0.05. Based on this, it was judged that the Y-shaped gap area of the endothelial cells with corneal endothelial decompensation was significantly higher than that of the endothelial cells with normal corneal endothelial function, further verifying that the Y-shaped gap area can effectively measure the corneal endothelial cell function. At the same time, 5 single-layer full corneal endothelial finite element models were constructed according to the corneal endothelial cell images of 5 individuals with normal corneal endothelial function, and the relative change magnifications of the Y-shaped gap areas of 5 normal corneal endothelial cells were calculated. The mean value of the relative change magnification data of the Y-shaped gap area was calculated to be 1.35, and the standard deviation was 0.12. The relative change magnification threshold of the Y-shaped gap area was constructed as 1.6. The relative change magnification threshold of the Y-shaped gap area was the mean value plus 2 times the standard deviation. In practical applications, the corneal endothelial cell image of a certain patient was collected and a model was constructed, and the relative change magnification of its Y-shaped gap area was obtained as 1.7, which was greater than the threshold value of 1.6. It was thus judged that the corneal endothelial cell function of this patient was abnormal. After subsequent clinical examinations, this patient was diagnosed with early corneal endothelial dysfunction, proving the accuracy and reliability of this evaluation method.

[0061] The above is the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of modifications or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. A method for evaluating the function of a single corneal endothelial cell based on cell morphology, characterized in that It includes the following steps: S1. Collect images of normal corneal endothelial cells through a corneal endothelial microscope, perform denoising and enhancement processing on the images, use the global threshold method to segment the corneal endothelial cell contours on the processed images, import the segmented corneal endothelial cell contours into the finite element analysis software ANSYS, construct a normal monolayer full corneal endothelial finite element model, and adjust the beam element parameters of the normal monolayer full corneal endothelial finite element model to generate a decompensated monolayer full corneal endothelial finite element model; S2. Apply pressures of 10 mmHg, 15 mmHg, and 20 mmHg to the normal and decompensated monolayer full corneal endothelial finite element models respectively for simulation analysis. Taking the parameter values of corneal endothelial cells including maximum displacement, edge stress, F / L ratio, and Y-shaped gap area under 10 mmHg pressure as reference values, calculate the relative change multiples of each parameter value under 15 mmHg and 20 mmHg pressure conditions. Based on the relative change multiples of each parameter value of corneal endothelial cells under the normal and decompensated monolayer full corneal endothelial finite element models, obtain the simulation results related to the Y-shaped gap area and corneal endothelial cell function; S3. Based on the simulation results, select 5 images of corneal endothelial cells with clinically diagnosed corneal endothelial function decompensation and normal respectively, construct a monolayer full corneal endothelial finite element model, calculate the Y-shaped gap area, and through independent sample t-test, it is obtained that the Y-shaped gap area of endothelial cells with corneal endothelial function decompensation is higher than that of endothelial cells with normal corneal endothelial function, and it is determined that the Y-shaped gap area can measure corneal endothelial cell function; S4. Determine the threshold of the relative change multiple of the Y-shaped gap area based on the relative change multiple of the Y-shaped gap area of normal corneal endothelial cells, collect images of corneal endothelial cells, construct a monolayer full corneal endothelial finite element model, obtain the relative change multiple of the Y-shaped gap area, and thereby judge the functional state of corneal endothelial cells.

2. The method for evaluating the function of a single corneal endothelial cell based on cell morphology according to claim 1, wherein The method for denoising and enhancing the image is as follows: Perform denoising processing on the image through non-local means denoising. Collect images of normal corneal endothelial cells through a corneal endothelial microscope. For each pixel in the image, find pixel blocks similar to the neighborhood window of the pixel within the set search window through the Euclidean distance calculation formula, perform weighted average processing on the pixel values used to describe the pixel gray characteristics in the pixel block to obtain the denoised pixel, and generate the denoised image for all pixels of the denoised image; Perform enhancement processing on the denoised image through contrast-limited adaptive histogram equalization (CLAHE). Divide the denoised image into non-overlapping sub-blocks according to the set block size. For each sub-block, adjust the pixel values through calculating the cumulative distribution function (CDF) to enhance the contrast, and use bilinear interpolation method to interpolate the edges of adjacent sub-blocks. Combine all sub-blocks together to obtain the enhanced image; The search window and neighborhood window are the regional ranges of the pixel block; The pixel block is an area composed of multiple adjacent pixels.

3. The method for evaluating the function of a single corneal endothelial cell based on cell morphology according to claim 2, wherein The method for segmenting the corneal endothelial cell contours using the global threshold method on the processed image is as follows: Read the processed image through the cv2.imread() function in Python. The processed image is stored in the computer memory in matrix form. Generate an array with 256 elements for the processed image in the computer memory through the cv2.calcHist() function of Python. Then, draw a grayscale histogram based on this through the matplotlib library of Python. Determine the global threshold through the Otsu algorithm according to the peaks and valleys of the grayscale histogram. Then traverse all the pixels of the processed image; Determine the pixels with gray values greater than the global threshold as corneal endothelial cell pixels and set the gray value to 0; Determine the pixels with gray values less than the global threshold as corneal endothelial cell contour pixels and set the gray value to 255; Obtain the segmented corneal endothelial cell contour accordingly; The grayscale histogram is a statistical chart showing the frequency of each gray level in a digital image; The Otsu algorithm is an automatic threshold determination algorithm.

4. The method for evaluating the function of a single corneal endothelial cell based on cell morphology according to claim 3, wherein, The method for generating a decompensated single-layer total corneal endothelium finite element model by adjusting the beam element parameters of a normal single-layer total corneal endothelium finite element model is as follows: The beam element parameters are elastic modulus and cross-sectional area; Based on the existing research data on the mechanical properties of corneal endothelial cell junction proteins, Determine the elastic modulus E0 and cross-sectional area A0 of normal corneal endothelial cell junction proteins. According to the characteristic of the decrease in the stiffness of junction proteins during decompensation, on the basis of the elastic modulus E0, adjust the elastic modulus in the beam element parameters of the normal single-layer total corneal endothelium finite element model. The adjustment range of the elastic modulus is from 0.2E0 to 0.5E0; According to the characteristic of the decrease in the number of junction proteins during decompensation, on the basis of the cross-sectional area A0, adjust the cross-sectional area in the beam element parameters of the normal single-layer total corneal endothelium finite element model. The adjustment range of the cross-sectional area is from 0.5A0 to 0.8A0.

5. The method for evaluating the function of a single corneal endothelial cell based on cell morphology according to claim 4, wherein The method for calculating the relative change magnification of each parameter value of corneal endothelial cells under 15 mmHg and 20 mmHg pressure conditions with the parameter values of corneal endothelial cells including the maximum displacement, edge stress, F / L ratio, and Y-shaped gap area under 10 mmHg pressure as the reference values is as follows: Using the finite element analysis software ANSYS, a pressure of 10 mmHg was applied to the normal and decompensated single-layer total corneal endothelium finite element models respectively, and the parameter values of the corneal endothelial cells such as the maximum displacement, edge stress, F / L ratio, and Y-shaped gap area were obtained. The parameter values were recorded as the reference values for calculating the relative change magnification. The maximum displacement of the normal single-layer total corneal endothelium finite element model under a pressure of 10 mmHg was D 10-ZC , the edge stress was S 10-ZC , the F / L ratio was R 10-ZC , the Y-shaped gap area was A 10-ZC , the maximum displacement of the decompensated single-layer total corneal endothelium finite element model under a pressure of 10 mmHg was D 10-SDC , the edge stress was S 10-SDC , the F / L ratio was R 10-SDC , the Y-shaped gap area was A 10-SDC ; By using the finite element analysis software ANSYS, the parameter values of the maximum displacement, edge stress, F / L ratio, and Y-shaped gap area of corneal endothelial cells of the normal and decompensated single-layer total corneal endothelium finite element models under pressures of 15 mmHg and 20 mmHg were obtained. The maximum displacement of the normal single-layer total corneal endothelium finite element model under a pressure of 15 mmHg was D 15-ZC , the edge stress was S 15-ZC , the F / L ratio was R 15-ZC , and the Y-shaped gap area was A 15-ZC . The maximum displacement of the decompensated single-layer total corneal endothelium finite element model under a pressure of 15 mmHg was D 15-SDC , the edge stress was S 15-SDC , the F / L ratio was R 15-SDC , and the Y-shaped gap area was A 15-SDC ; The maximum displacement of the normal single-layer full corneal endothelium finite element model under a pressure of 20 mmHg is D 20-ZC , the edge stress is S 20-ZC , the F / L ratio is R 20-ZC , the Y-shaped gap area is A 20-ZC , the maximum displacement of the decompensated single-layer full corneal endothelium finite element model under a pressure of 20 mmHg is D 20-SDC , the edge stress is S 20-SDC , the F / L ratio is R 20-SDC , the Y-shaped gap area is A 20-SDC ; Calculate the relative change magnification of each parameter value through the relative change magnification calculation formula for the parameter values under 15 mmHg and 20 mmHg pressure conditions.

6. The method for evaluating the function of a single corneal endothelial cell based on cell morphology according to claim 5, wherein The method for obtaining the correlation between the Y-shaped gap area and the function of corneal endothelial cells according to the relative change magnification of each parameter value of corneal endothelial cells under normal and decompensated single-layer total corneal endothelium finite element models is as follows: Divide the relative change magnification data of each parameter value of normal and decompensated single-layer total corneal endothelium finite element models under 15 mmHg and 20 mmHg pressure into a normal group and a decompensated group. The normal group includes the relative change magnification of each parameter value of the normal single-layer total corneal endothelium finite element model under 15 mmHg and 20 mmHg pressure, and the decompensated group includes the relative change magnification of each parameter value of the decompensated single-layer total corneal endothelium finite element model under 15 mmHg and 20 mmHg pressure; Analyze the trend of the relative change multiples of each parameter value with the change in pressure in the normal group and the decompensated group, and find the parameter value with the largest increase in the relative change multiple at 15 mmHg and 20 mmHg pressures. At the same time, if the relative change multiple of the parameter value with the largest increase in the relative change multiple in the decompensated group is higher than the relative change multiple in the corresponding normal group, it is concluded that this parameter value is related to the corneal endothelial cell function. Based on this, it is concluded that the Y-shaped gap area is related to the corneal endothelial cell function.

7. The method for evaluating the function of a single corneal endothelial cell based on cell morphology according to claim 6, characterized in that, The method to determine that the Y-shaped gap area of the endothelial cells with corneal endothelial dysfunction is higher than that of the endothelial cells with normal corneal endothelial function through independent samples t-test is as follows: The relative change multiples of the Y-shaped gap area corresponding to the images of corneal endothelial cells with corneal endothelial decompensation and normal function are statistically analyzed using the independent samples t-test method respectively. If the P value corresponding to the Y-shaped gap area is less than the pre-set 0.05, it is determined that the Y-shaped gap area can measure the corneal endothelial cell function; The independent samples t-test is a statistical method used to compare whether the difference in the means of two independent samples is significant.

8. A method for evaluating the function of a single corneal endothelial cell based on cell morphology according to claim 7, characterized in that, The method to determine the threshold of the relative change multiple of the Y-shaped gap area based on the relative change multiple of the Y-shaped gap area of normal corneal endothelial cells is as follows: Collect 10 image samples of clinically normal corneal endothelial cells, construct 10 normal single-layer full corneal endothelial finite element model samples, and thus obtain the relative change multiples of the Y-shaped gap area of 10 normal corneal endothelial cells. Calculate the mean and standard deviation of the relative change multiple data for the relative change multiples of the Y-shaped gap area of 10 normal corneal endothelial cells, and thus construct the threshold of the relative change multiple of the Y-shaped gap area.

9. The method for evaluating the function of a single corneal endothelial cell based on cell morphology according to claim 8, characterized in that The method to judge the functional state of corneal endothelial cells is as follows: Compare the relative change multiple of the Y-shaped gap area of corneal endothelial cells with the threshold of the relative change multiple of the Y-shaped gap area; If the relative change multiple of the Y-shaped gap area of corneal endothelial cells is less than the threshold of the relative change multiple of the Y-shaped gap area, it is judged that the corneal endothelial cell function is in a normal state; If the relative change multiple of the Y-shaped gap area of corneal endothelial cells is greater than or equal to the threshold of the relative change multiple of the Y-shaped gap area, it is judged that the corneal endothelial cell function is abnormal.

Images